Claude Code · Codex · Cursor · OpenCode
Design Skills for Claude Code, Codex and Your Agent.
244 design skills for slides, interfaces, brand systems, motion, research and more — catalogued from real SKILL.md files. Start with the job you need to finish, then check the required tools and license.
01 All skills
All 244 Design Skills, by Task.
Compare the task, output, license and required tools. Each card links to its upstream SKILL.md; repository stars are not skill install counts.
-
Break UI
by Emil Kowalski · emilkowalski/skills
Stress-test existing UI with realistic extreme data and a dev-only comparison toggle; report failures before fixing them.
Component source, existing demo data and field limits; a running app and browser for visual verification.
-
Animation Performance
by Emil Kowalski · emilkowalski/skills
Diagnose dropped frames in web animations: rendering cost, main-thread contention, per-frame React updates and inherited CSS variables.
Animation source and a reproducible interaction; browser profiling and a real target device to verify performance.
-
Image to UI
by UIZZE · uizze/uizze
Recreate a supplied screenshot in the existing app, then render at the reference viewport and correct visible differences.
Provide a readable reference image, its target viewport, project access and available assets. Browser rendering is needed to verify the visual match. No UIZZE account or MCP connection is required.
-
HyperFrames
by HeyGen · heygen-com/hyperframes
Route video requests to the right workflow, resume existing projects and coordinate HTML composition, media, preview and rendering.
The CLI requires Node.js 22+ and FFmpeg. Start with the core skills; creation workflows are installed on demand. Media services and cloud rendering may add requirements.
-
Anatomy
by wheresryan22 · wheresryan22/anatomy
Explain technical ideas through crafted isometric SVG mechanisms and model-driven interaction.
Node.js 20.10+ for the kit; Chrome or Chromium for checks. The skills CLI has a separate Node requirement.
-
algorithmic-art
by Anthropic · anthropics/skills
Generative art with p5.js: seeded particles, flow fields, and tunable parameters.
Needs: p5.js
-
brand-guidelines
by Anthropic · anthropics/skills
Applies Anthropic's official brand colors and typography to any artifact. Anthropic brand only.
-
canvas-design
by Anthropic · anthropics/skills
Original posters and static visual pieces delivered as PNG or PDF.
-
frontend-design
by Anthropic · anthropics/skills
Sets a distinctive visual direction, typography, and layout for new or existing web UI.
-
slack-gif-creator
by Anthropic · anthropics/skills
Animated GIFs that meet Slack's size and dimension limits.
-
theme-factory
by Anthropic · anthropics/skills
Applies or generates color-and-type themes for pages, slides, and documents.
-
web-artifacts-builder
by Anthropic · anthropics/skills
Multi-component interactive HTML artifacts with React, Tailwind, and shadcn/ui.
-
design-consultation
by Garry Tan · garrytan/gstack
Product design consultation that proposes a full design system: aesthetic, type, color, layout, motion.
-
design-html
by Garry Tan · garrytan/gstack
Builds HTML/CSS from an agreed design direction, with project-specific review before release.
-
design-review
by Garry Tan · garrytan/gstack
Designer's-eye QA of a built page: inconsistency, spacing, hierarchy, slop patterns; then fixes them.
-
design-shotgun
by Garry Tan · garrytan/gstack
Generates several design variants in parallel and opens a comparison board for feedback.
-
diagram
by Garry Tan · garrytan/gstack
Creates Mermaid source, Excalidraw, SVG and PNG; element editing depends on the diagram type.
Needs: Excalidraw
-
ios-design-review
by Garry Tan · garrytan/gstack
Visual design audit of iOS apps on real hardware.
-
plan-design-review
by Garry Tan · garrytan/gstack
Reviews a design plan for completeness and experience risk before implementation.
-
banner-design
by nextlevelbuilder · nextlevelbuilder/ui-ux-pro-max-skill
Banners for social media, ads, website heroes, and print.
-
brand
by nextlevelbuilder · nextlevelbuilder/ui-ux-pro-max-skill
Brand voice, visual identity, messaging frameworks, and brand consistency.
-
design
by nextlevelbuilder · nextlevelbuilder/ui-ux-pro-max-skill
Brand identity, design tokens, UI styling, logo generation, and corporate identity programs.
Needs: Image generation service (Gemini, Atlas Cloud, or MuAPI)
-
design-system
by nextlevelbuilder · nextlevelbuilder/ui-ux-pro-max-skill
Three-tier design tokens, component specifications, and slide generation.
-
slides
by nextlevelbuilder · nextlevelbuilder/ui-ux-pro-max-skill
Strategic HTML slide decks with Chart.js, design tokens, and responsive layouts.
Needs: Chart.js
-
ui-styling
by nextlevelbuilder · nextlevelbuilder/ui-ux-pro-max-skill
React UI styling with shadcn/ui, Tailwind and a separate canvas-design reference.
Choose the component, Tailwind-only or canvas path; check project versions and companion references.
-
ui-ux-pro-max
by nextlevelbuilder · nextlevelbuilder/ui-ux-pro-max-skill
UI/UX design intelligence for web, mobile, and desktop: layout, color, type, and UX rules.
-
shadcn
by shadcn · shadcn-ui/ui
Adds, searches, fixes, styles, and composes shadcn components in a project.
Needs: shadcn/ui
-
frontend-ui-engineering
by Addy Osmani · addyosmani/agent-skills
Builds production-quality, accessible, responsive user-facing UIs.
-
brandkit
by Leonxlnx · Leonxlnx/taste-skill
Generates brand-identity concept boards as images, including logo, color and typography directions.
Needs: Image generation service
-
design-taste-frontend
by Leonxlnx · Leonxlnx/taste-skill
Anti-slop frontend design for landing pages, portfolios, and redesigns.
-
gpt-taste
by Leonxlnx · Leonxlnx/taste-skill
Editorial layouts, layout variety, and GSAP motion in one frontend skill.
Needs: GSAP
-
high-end-visual-design
by Leonxlnx · Leonxlnx/taste-skill
Teaches high-end agency detail: type, spacing, shadows, cards, and motion.
-
image-to-code
by Leonxlnx · Leonxlnx/taste-skill
Generates and analyzes a design image first, then implements the matching page.
Needs: Image generation service
-
imagegen-frontend-mobile
by Leonxlnx · Leonxlnx/taste-skill
High-fidelity mobile screen concepts and flows via image generation.
Needs: Image generation service
-
imagegen-frontend-web
by Leonxlnx · Leonxlnx/taste-skill
Section-by-section website design references via image generation.
Needs: Image generation service
-
industrial-brutalist-ui
by Leonxlnx · Leonxlnx/taste-skill
Brutalist interfaces fusing Swiss typography with terminal aesthetics.
-
minimalist-ui
by Leonxlnx · Leonxlnx/taste-skill
Clean, monochrome, editorial-style interfaces.
-
redesign-existing-projects
by Leonxlnx · Leonxlnx/taste-skill
Audits an existing interface, implements selected redesign changes and verifies the result.
-
stitch-design-taste
by Leonxlnx · Leonxlnx/taste-skill
Writes a DESIGN.md for Google Stitch with an emphasis on visual distinctiveness.
Needs: Google Stitch
-
impeccable
by Paul Bakaus · pbakaus/impeccable
One skill to design, critique, audit, polish, animate, and refine interfaces.
-
faceless-explainer
by HeyGen · heygen-com/hyperframes
Turns text, notes, or an article into a faceless explainer video with invented scene visuals.
Needs: HyperFrames
-
figma
by HeyGen · heygen-com/hyperframes
Imports Figma assets, brand tokens, and storyboards into a HyperFrames composition.
Needs: HyperFrames, Figma
-
general-video
by HeyGen · heygen-com/hyperframes
Authors or edits a custom multi-scene HyperFrames video when no template fits.
Needs: HyperFrames
-
hyperframes-animation
by HeyGen · heygen-com/hyperframes
Animation rules, scene blueprints, transitions, and runtime adapters for HyperFrames.
Needs: HyperFrames
-
hyperframes-creative
by HeyGen · heygen-com/hyperframes
Creative direction for HyperFrames videos: palette, type, composition, voice-over, pacing.
Needs: HyperFrames
-
hyperframes-keyframes
by HeyGen · heygen-com/hyperframes
Punch-ins, zooms, reframes, Ken Burns moves, and 2D/3D keyframes for HyperFrames.
Needs: HyperFrames
-
motion-graphics
by HeyGen · heygen-com/hyperframes
Short design-led motion graphics: kinetic type, stat count-ups, logo stings, lower thirds.
Needs: HyperFrames
-
product-launch-video
by HeyGen · heygen-com/hyperframes
Turns a product URL, script, or brief into a product launch or promo video.
Needs: HyperFrames
-
slideshow
by HeyGen · heygen-com/hyperframes
Interactive HyperFrames slide shows with fragments, branching, and presenter mode.
Needs: HyperFrames
-
slidev
by Slidev · slidevjs/slidev
Developer slide decks with Markdown, Vue components, code highlighting, and animation.
Needs: Slidev
-
Adobe Illustrator Scripting
by GitHub Awesome Copilot Contributors · github/awesome-copilot
Write ExtendScript scripts to create and edit Illustrator documents, artwork and exports.
Adobe Illustrator CC or later; generated scripts execute in its ExtendScript environment.
-
Anti UI Slop
by UIZZE · github/awesome-copilot
Start with the product's real objects and constraints before choosing a visual treatment.
An agent able to read project files; rendering support for visual checks. Paid UIZZE MCP is optional.
-
FreeCAD Scripts
by GitHub Awesome Copilot Contributors · github/awesome-copilot
Translate a dimensioned modeling task into FreeCAD objects that remain inspectable and adjustable.
FreeCAD with its bundled Python; GUI work additionally uses bundled PySide and scenegraph work uses Pivy.
-
Generate Image
by GitHub Awesome Copilot Contributors · github/awesome-copilot
Generate and save image assets through a supported provider, using the project's configured credentials.
A supported provider API key and network access; model availability and usage charges must be checked separately.
-
Nano Banana Pro OpenRouter
by GitHub Awesome Copilot Contributors · github/awesome-copilot
Generate or edit images through OpenRouter using reference images and a bundled Python script.
uv, the bundled Python script and an OPENROUTER_API_KEY with appropriate model access and credits.
-
Premium Frontend UI
by Utkarsh Patrikar · github/awesome-copilot
A strongly opinionated frontend guide for expressive landing pages, portfolios and interactive presentations.
A frontend project and any deliberately selected animation libraries; no single runtime is required by the guide.
-
Rhino 3D Scripts
by GitHub Awesome Copilot Contributors · github/awesome-copilot
Choose a macro for a command sequence, and a script when the modeling task needs logic.
Rhino 7 or later; Rhino 8 is the source's recommended environment for the unified Script Editor.
-
Slang Shader Engineer
by GitHub Awesome Copilot Contributors · github/awesome-copilot
Write Slang shaders and check how their inputs, resources and outputs connect to the renderer.
A Slang compilation and renderer environment appropriate to the target backend; host-side validation is needed for integration claims.
-
accessibility-compliance
by Seth Hobson · wshobson/agents
WCAG 2.2 compliant interfaces with mobile accessibility and assistive-technology support.
-
brand-landingpage
by Seth Hobson · wshobson/agents
Brand-identity interview first, then generates and iterates a landing page through Stitch.
Needs: Google Stitch
-
design-system-patterns
by Seth Hobson · wshobson/agents
Scalable design systems: tokens, theming infrastructure, component architecture.
-
interaction-design
by Seth Hobson · wshobson/agents
Implements microinteractions, transitions, loading states and user-feedback patterns.
-
kpi-dashboard-design
by Seth Hobson · wshobson/agents
KPI dashboards: metric selection, chart choice, layout, and real-time patterns.
-
mobile-android-design
by Seth Hobson · wshobson/agents
Native Android UI with Material Design 3 and Jetpack Compose.
-
mobile-ios-design
by Seth Hobson · wshobson/agents
Native iOS UI following Apple's Human Interface Guidelines with SwiftUI.
-
pptx-quality-gates
by Seth Hobson · wshobson/agents
Validates and repairs editable PPTX decks for geometry, accessibility, and package integrity.
-
pptx-reference-deck-analysis
by Seth Hobson · wshobson/agents
Analyzes a reference PPTX for theme, typography, layout rhythm, and diagnostics.
-
pptx-slide-specification
by Seth Hobson · wshobson/agents
Authors coordinate-explicit JSON specifications for editable PPTX decks.
-
pptx-visual-assets
by Seth Hobson · wshobson/agents
Selects and places approved icons, images, SVGs, and infographics in a PPTX deck.
-
react-native-design
by Seth Hobson · wshobson/agents
React Native styling, navigation, and Reanimated animations.
Needs: React Native
-
responsive-design
by Seth Hobson · wshobson/agents
Responsive layouts with container queries, fluid type, CSS Grid, and mobile-first breakpoints.
-
screen-reader-testing
by Seth Hobson · wshobson/agents
Guides real screen-reader checks of web apps with VoiceOver, NVDA and other readers.
Needs a runnable app, real screen readers and supported test platforms.
-
tailwind-design-system
by Seth Hobson · wshobson/agents
Design systems with Tailwind CSS v4: tokens, component libraries, responsive patterns.
Needs: Tailwind CSS
-
visual-design-foundations
by Seth Hobson · wshobson/agents
Typography, color theory, spacing systems, and iconography for cohesive visuals.
-
wcag-audit-patterns
by Seth Hobson · wshobson/agents
WCAG 2.2 audits with automated tests, manual checks, and remediation guidance.
-
web-component-design
by Seth Hobson · wshobson/agents
Reusable React, Vue, and Svelte component patterns and composition strategies.
-
draw-io-diagram-generator
by GitHub · github/awesome-copilot
Creates and edits draw.io diagram files (.drawio, .drawio.svg, .drawio.png).
Needs: draw.io
-
excalidraw-diagram-generator
by GitHub · github/awesome-copilot
Generates Excalidraw diagrams and mind maps from natural language.
Needs: Excalidraw
-
gsap-framer-scroll-animation
by GitHub · github/awesome-copilot
Parallax, pinned sections, horizontal scroll, and text scroll animations.
Needs: GSAP
-
penpot-uiux-design
by GitHub · github/awesome-copilot
Professional UI/UX design, components, and design systems in Penpot through MCP tools.
Needs: Penpot MCP
-
power-bi-report-design-consultation
by GitHub · github/awesome-copilot
Power BI report design: chart selection, layout, readability, accessibility.
Needs: Power BI
-
web-design-reviewer
by GitHub · github/awesome-copilot
Visually inspects a running website and locates and fixes layout issues.
-
animate
by Emil Kowalski · emilkowalski/skills
Builds an animation from scratch, deciding purpose, tool, properties, curve, and duration in order.
-
animate-expo
by Emil Kowalski · emilkowalski/skills
Gestures, springs, and transitions for React Native and Expo.
Needs: React Native / Expo
-
animation-vocabulary
by Emil Kowalski · emilkowalski/skills
Reverse-lookup glossary that turns a vague motion description into its exact term.
-
apple-design
by Emil Kowalski · emilkowalski/skills
Adapts Apple-inspired gestures, physical motion and spatial hierarchy to web interfaces.
-
emil-design-eng
by Emil Kowalski · emilkowalski/skills
Design-engineering guidance for polished web interfaces, interaction states and frontend review.
-
find-animation-opportunities
by Emil Kowalski · emilkowalski/skills
Finds places in a UI that should animate, with parameter suggestions, and rejects the rest.
-
improve-animations
by Emil Kowalski · emilkowalski/skills
Audits existing motion code and produces prioritized, self-contained implementation plans.
-
prototype
by Emil Kowalski · emilkowalski/skills
Builds several genuinely different versions of a UI piece behind a live visual picker.
-
review-animations
by Emil Kowalski · emilkowalski/skills
Reviews animation code for timing, interaction, and finish against a high craft bar.
-
vercel-react-view-transitions
by Vercel Labs · vercel-labs/agent-skills
Native-feeling page transitions with React's View Transition API.
Needs: React
-
frontend-slides
by Zara Zhang · zarazhangrui/frontend-slides
Animation-rich HTML slides from scratch or converted from PowerPoint (PPT) files.
-
Figma Code Connect Components
by OpenAI · openai/skills
Map published Figma components to their matching code components with Code Connect.
Connected Figma MCP, access to published components and the codebase, plus current Code Connect entitlement and edit permissions.
-
Figma Create Design System Rules
by OpenAI · openai/skills
Document the repository paths, components and conventions an agent should use to implement Figma designs.
Figma MCP connection and access to the target codebase; the destination rule format depends on the coding host.
-
Figma Generate Design
by OpenAI · openai/skills
Create an editable Figma screen that stays connected to its design system.
Connected Figma MCP, target-file access, a published design system and the companion figma-use Skill for canvas operations.
-
Figma Generate Library
by OpenAI · openai/skills
Build a reusable Figma library, starting with design tokens and then adding components.
Figma MCP with editable-file access, repository access and the companion figma-use Skill plus helper references.
-
Figma Implement Design
by OpenAI · openai/skills
Implement a Figma design using the target repository's components and conventions, then compare the result with the design.
Connected Figma MCP, access to the selected design and the target repository, plus a runnable preview for visual validation.
-
ink
by Google Labs · google-labs-code/design.md
Renders JSON specs into interactive terminal UIs with Ink.
Needs: Ink
-
Accessibility Audit
by Alireza Rezvani · alirezarezvani/claude-skills
Keep each accessibility finding connected to the affected element, proposed fix and verification result.
Python for the bundled tools, source access and a browser or assistive-technology setup for contextual validation.
-
Apple HIG Expert
by Alireza Rezvani · alirezarezvani/claude-skills
Review Apple-platform interfaces against platform-specific design and accessibility guidance.
Python 3 for the checker, which uses only the standard library; current Apple Human Interface Guidelines (HIG); and device or simulator access for behavior checks.
-
Landing Page Generator
by Alireza Rezvani · alirezarezvani/claude-skills
Connect a real product proposition to page sections before generating the components.
A compatible React or Next.js project with Tailwind; referenced UI components and optional brand-analysis tooling may require separate setup.
-
imagegen
by OpenAI · openai/skills
Generates or edits raster images: illustrations, photos, textures, sprites, mockups.
Needs: Image generation service
-
Baoyu Article Illustrator
by Jim Liu (Baoyu) · JimLiu/baoyu-skills
Plan and generate illustrations that explain key ideas at relevant points in an article.
An available raster-generation backend and the Skill's reference/configuration files; backend access and cost are separate.
-
Baoyu Comic
by Jim Liu (Baoyu) · JimLiu/baoyu-skills
Plan what each panel teaches before generating the comic pages.
An available image backend; Bun or an npx-based Bun runtime for the supplied PDF merge script.
-
Baoyu Cover Image
by Jim Liu (Baoyu) · JimLiu/baoyu-skills
Generate an article cover around one central idea, with a defined composition and visual style.
An available raster image backend and the Skill's configuration and art-direction references.
-
Baoyu Diagram
by Jim Liu (Baoyu) · JimLiu/baoyu-skills
Keep relationships, labels and connector paths explicit in an inspectable vector file.
A text/code environment for SVG; Bun or npx-based Bun for the supplied PNG conversion script.
-
Baoyu Image Gen
by Jim Liu (Baoyu) · JimLiu/baoyu-skills
Run image generation and editing jobs through supported providers, including batch workflows.
Bun or an npx-based Bun runtime, complete provider adapters and credentials or login for the chosen supported route.
-
Baoyu Infographic
by Jim Liu (Baoyu) · JimLiu/baoyu-skills
Turn content into an infographic using a comparison, hierarchy, process or overview layout suited to the material.
An available raster image backend plus layout, style and configuration references.
-
Baoyu Slide Deck
by Jim Liu (Baoyu) · JimLiu/baoyu-skills
Generate illustrated slides and assemble an image-based deck for sharing; individual slide objects are not editable.
A raster generation backend and Bun or npx-based Bun for PPTX/PDF assembly scripts.
-
Baoyu XHS Images
by Jim Liu (Baoyu) · JimLiu/baoyu-skills
Create a visually consistent series of Xiaohongshu image cards, with one main idea per card.
An available raster backend with suitable reference support, plus configuration and card-layout references.
-
guizang-ppt-skill
by op7418 · op7418/guizang-ppt-skill
Creates horizontal HTML decks with speaker view and sync, plus supporting assets and external resources.
-
brand-guidelines
by Alireza Rezvani · alirezarezvani/claude-skills
Defines or applies brand guidelines for any product: color, type, logo, and voice.
-
design-system
by Alireza Rezvani · alirezarezvani/claude-skills
Captures brand identity once through a 10-question wizard for later documents and decks.
-
md-slides
by Alireza Rezvani · alirezarezvani/claude-skills
Converts a Markdown deck into branded slides.
-
ui-design-system
by Alireza Rezvani · alirezarezvani/claude-skills
Design tokens, component documentation, responsive calculations, and developer handoff.
-
ux-researcher-designer
by Alireza Rezvani · alirezarezvani/claude-skills
Data-driven personas, journey maps, usability testing frameworks, and research synthesis.
-
huashu-design
by alchaincyf · alchaincyf/huashu-design
High-fidelity HTML prototypes, slides, animation, and visualizations; starts from three drafts.
-
accessibility-review
by Anthropic · anthropics/knowledge-work-plugins
WCAG 2.1 AA audit of a design or page: contrast, keyboard, touch targets.
-
design-critique
by Anthropic · anthropics/knowledge-work-plugins
Structured design feedback on usability, hierarchy, and consistency.
-
design-handoff
by Anthropic · anthropics/knowledge-work-plugins
Developer handoff specs from a design: layout, tokens, component states, breakpoints.
-
design-system
by Anthropic · anthropics/knowledge-work-plugins
Audits, documents, or extends an existing design system.
-
research-synthesis
by Anthropic · anthropics/knowledge-work-plugins
Synthesizes interviews, surveys, and test notes into themes and recommendations.
-
user-research
by Anthropic · anthropics/knowledge-work-plugins
Plans, conducts, and synthesizes user interviews, surveys, and usability tests.
-
ux-copy
by Anthropic · anthropics/knowledge-work-plugins
Writes interface microcopy, error messages, empty states and action labels.
-
gsap-core
by GreenSock · greensock/gsap-skills
GSAP core API: to/from/fromTo, easing, stagger, matchMedia, reduced motion.
Needs: GSAP
-
gsap-frameworks
by GreenSock · greensock/gsap-skills
GSAP in Vue, Svelte, and other non-React frameworks: lifecycle, scoping, cleanup.
Needs: GSAP
-
gsap-performance
by GreenSock · greensock/gsap-skills
GSAP performance: transforms, avoiding layout thrash, will-change, batching.
Needs: GSAP
-
gsap-plugins
by GreenSock · greensock/gsap-skills
GSAP plugins: ScrollTo, ScrollSmoother, Flip, Draggable, SplitText, SVG and physics.
Needs: GSAP
-
gsap-react
by GreenSock · greensock/gsap-skills
GSAP in React and Next.js with the useGSAP hook, refs, context, and cleanup.
Needs: GSAP
-
gsap-scrolltrigger
by GreenSock · greensock/gsap-skills
Scroll-linked animation with ScrollTrigger: pinning, scrub, triggers, parallax.
Needs: GSAP
-
gsap-timeline
by GreenSock · greensock/gsap-skills
GSAP timelines: position parameter, nesting, and playback control.
Needs: GSAP
-
gsap-utils
by GreenSock · greensock/gsap-skills
gsap.utils helpers: clamp, mapRange, interpolate, snap, wrap, pipe.
Needs: GSAP
-
color-font-skill
by MiniMax · MiniMax-AI/skills
Presentation-ready color palettes and font pairings for PPT and design tasks.
-
design-style-skill
by MiniMax · MiniMax-AI/skills
A consistent visual system for PPT slides from radius and spacing recipes.
-
gif-sticker-maker
by MiniMax · MiniMax-AI/skills
Turns photos of people, pets, objects, or logos into four captioned GIF stickers.
-
pptx-generator
by MiniMax · MiniMax-AI/skills
Generates, edits, and reads editable PowerPoint (PPTX) slide decks.
-
shader-dev
by MiniMax · MiniMax-AI/skills
GLSL shaders: ray marching, SDFs, fluids, particles, procedural visuals.
Needs: GLSL / WebGL
-
slide-making-skill
by MiniMax · MiniMax-AI/skills
Implements single PowerPoint slides with PptxGenJS.
Needs: PptxGenJS
-
Beautiful Article
by ConardLi · ConardLi/garden-skills
Turn source material into a readable HTML article with editorial typography, layout and illustrations.
The documented build uses Node/npm, Vite, React, TypeScript, and Reacticle. Document extraction and optional PDF output may require additional tools; generated imagery needs a separately available backend.
-
GPT Image 2
by ConardLi · ConardLi/garden-skills
Prepare image prompts and run generation or editing through an available native image tool or the configured local API workflow.
Local API mode requires explicit Garden enablement, Node scripts, and configured credentials. Native mode depends on the host image tool. Model access, pricing, and supported parameters must be checked in that environment.
-
Web Design Engineer
by ConardLi · ConardLi/garden-skills
Establish a visual direction, build a representative interface and review it in the browser before expanding the design.
Needs a code-capable agent and browser-renderable frontend environment. Selected patterns may use React, Babel, charts, or other libraries; browser acceptance requires a suitable browser tool.
-
Web Video Presentation
by ConardLi · ConardLi/garden-skills
Build browser-based presentations with timed visual scenes and optional narration; recording is a separate step.
Uses Vite, React, TypeScript, and Node tooling. Optional TTS depends on the selected provider, credentials, and adapter; final recording needs browser capture or recording software.
-
Video Shotcraft
by Vincentwei1021 · Vincentwei1021/video-shotcraft
Plan and assemble a Remotion product video from interface captures, shot instructions and a timed composition.
Requires a Remotion/Node environment, browser capture, and ffmpeg. Some components need Three.js-related packages; optional music analysis and Jianying export add separate dependencies. Review Remotion's applicable license independently.
-
Scroll World
by oso95 · oso95/scroll-world
A pre-rendered camera journey controlled by scrolling, not a live 3D world.
The documented pipeline uses Higgsfield and a Monid video path, plus ffmpeg/ffprobe and a browser environment. Provider accounts, credits, model capabilities, and any optional image route require fresh verification.
-
Dashi PPT
by chuspeeism · chuspeeism/dashi-ppt-skill
Generate alternative slide layouts from one content source, then use the local editor to preview and export them.
Requires Node.js 20+, npm, and the generator packaged with the skill. Use its local preview service for editing/export features; a generic static server does not provide those APIs.
-
baseline-ui
by Julien Thibeaut · ibelick/ui-skills
Quickly de-slops UI code by fixing spacing, hierarchy, typography, and small layout issues.
-
create-design-md
by Julien Thibeaut · ibelick/ui-skills
Creates or updates a DESIGN.md from an existing product repository or public website.
-
fixing-accessibility
by Julien Thibeaut · ibelick/ui-skills
Audits and fixes ARIA, keyboard, focus, contrast, and form errors.
-
fixing-motion-performance
by Julien Thibeaut · ibelick/ui-skills
Finds and fixes janky CSS/JS animation and scroll-linked motion.
-
improve-ui
by Julien Thibeaut · ibelick/ui-skills
Audits a product surface against its own design evidence and writes implementation plans.
-
design-md
by Google Labs · google-labs-code/stitch-skills
Analyzes Stitch projects and writes a semantic DESIGN.md.
Needs: Google Stitch
-
enhance-prompt
by Google Labs · google-labs-code/stitch-skills
Rewrites vague UI ideas into polished, Stitch-optimized prompts.
Needs: Google Stitch
-
react-vite-dashboard
by Google Labs · google-labs-code/stitch-skills
Converts Stitch designs into React + Vite dashboards with TanStack Query.
Needs: Google Stitch
-
remotion
by Google Labs · google-labs-code/stitch-skills
Walkthrough videos from Stitch projects with Remotion transitions and zooms.
Needs: Google Stitch, Remotion
-
shadcn-ui
by Google Labs · google-labs-code/stitch-skills
Integrating and customizing shadcn/ui components: discovery, install, best practices.
Needs: shadcn/ui
-
stitch::code-to-design
by Google Labs · google-labs-code/stitch-skills
Converts frontend code into a Stitch design via static HTML and design-system extraction.
Needs: Google Stitch
-
stitch::extract-design-md
by Google Labs · google-labs-code/stitch-skills
Extracts a complete DESIGN.md from frontend source code in any framework.
-
stitch::extract-static-html
by Google Labs · google-labs-code/stitch-skills
Extracts static HTML from a running page; remaining external assets and dynamic behavior need checking.
-
stitch::generate-design
by Google Labs · google-labs-code/stitch-skills
Generates, edits, and varies screens in Stitch from text or reference images.
Needs: Google Stitch
-
stitch::manage-design-system
by Google Labs · google-labs-code/stitch-skills
Creates, updates, and applies design systems in Stitch through MCP tools.
Needs: Google Stitch
-
stitch::react-components
by Google Labs · google-labs-code/stitch-skills
Converts or syncs Stitch designs to modular React + Vite components.
Needs: Google Stitch
-
stitch::react-native
by Google Labs · google-labs-code/stitch-skills
Converts or syncs Stitch HTML designs to React Native components.
Needs: Google Stitch, React Native
-
App Store Screenshots
by ParthJadhav · ParthJadhav/app-store-screenshots
Compose a coordinated set of app-store listing images from screenshots, captions and visual assets.
Requires the packaged Next.js editor, a supported Node/package-manager environment, and a browser. PNG export uses html-to-image; source captures, fonts, and other assets need appropriate usage rights.
-
Codex PPT
by ningzimu · ningzimu/codex-ppt-skill
Turn source material into illustrated slides and assemble a PPTX with image-based, rather than individually editable, slide content.
Python 3, an available image-generation backend, and the repository's assembly and state-recording scripts. Referenced workflow documents are required for actual execution.
-
interface-design
by Dammyjay93 · Dammyjay93/interface-design
Craft-first interface design for dashboards, admin panels, SaaS tools, and settings pages.
-
Remotion Captions
by remotion-dev · remotion-dev/skills
Keep subtitle text and timing in a defined data structure before styling it on a video timeline.
A Remotion project, @remotion/captions, and the relevant transcription/import/display reference files.
-
Remotion Create
by remotion-dev · remotion-dev/skills
Set up a Remotion project and composition from a video brief, with a preview available during development.
Node.js, Git, package installation, Remotion Studio, and the related composition/layout references.
-
Remotion Interactivity
by remotion-dev · remotion-dev/skills
Make a video editable in Studio by changing how its React source exposes layers, timing, and controls.
A Remotion version supporting the documented Interactive APIs; @remotion/paths is additionally needed for the described editable path interpolation.
-
Remotion Maps
by remotion-dev · remotion-dev/skills
Build animated maps in Remotion, choosing the mapping technique, provider and data for the intended shot.
Remotion plus the selected technique's libraries and data; Mapbox requires a key according to the captured source.
-
Remotion Markup
by remotion-dev · remotion-dev/skills
Write React markup that behaves like a video timeline rather than a webpage with CSS animation.
Remotion and feature-specific packages matching the project version; linked references determine additional media, effects, font, or export requirements.
-
Baoyu Design
by JimLiu · JimLiu/baoyu-design
Use a file-based design workflow for prototypes, presentation concepts, and reusable visual systems.
A file-capable agent and the repository's methodology, built-in guides, reference documents, and scripts. Export and import paths have their own prerequisites.
-
Three.js Animation
by CloudAI-X · CloudAI-X/threejs-skills
Give a loaded character or product model a defined playback and blending strategy.
A working Three.js project and renderer; verify source examples against the installed release. Additional controls and loaders depend on the chosen path.
-
Three.js Geometry
by CloudAI-X · CloudAI-X/threejs-skills
Choose a shape representation before adding surface styling or animation.
A working Three.js project and renderer; verify source examples against the installed release. Additional controls and loaders depend on the chosen path.
-
Three.js Interaction
by CloudAI-X · CloudAI-X/threejs-skills
Connect a gesture to the intended 3D object using the canvas and camera's actual coordinates.
A working Three.js project and renderer; verify source examples against the installed release. Additional controls and loaders depend on the chosen path.
-
Three.js Lighting
by CloudAI-X · CloudAI-X/threejs-skills
Make a model readable with a deliberate lighting setup rather than indiscriminate brightness.
A working Three.js project and renderer; verify source examples against the installed release. Additional controls and loaders depend on the chosen path.
-
Three.js Materials
by CloudAI-X · CloudAI-X/threejs-skills
Match the shading model to the surface, then inspect it under the intended illumination.
A working Three.js project and renderer; verify source examples against the installed release. Additional controls and loaders depend on the chosen path.
-
Three.js Post-Processing
by CloudAI-X · CloudAI-X/threejs-skills
Treat screen effects as an ordered rendering pipeline instead of unrelated filters.
A working Three.js project and renderer; verify source examples against the installed release. Additional controls and loaders depend on the chosen path.
-
Three.js Shaders
by CloudAI-X · CloudAI-X/threejs-skills
Use custom vertex or pixel behavior when material presets cannot express the effect.
A working Three.js project and renderer; verify source examples against the installed release. Additional controls and loaders depend on the chosen path.
-
Three.js Textures
by CloudAI-X · CloudAI-X/threejs-skills
Assign each image its correct role before diagnosing a stretched or washed-out surface.
A working Three.js project and renderer; verify source examples against the installed release. Additional controls and loaders depend on the chosen path.
-
make-interfaces-feel-better
by Jakub Krehel · jakubkrehel/make-interfaces-feel-better
Polishes interface typography, surfaces, icons and microinteractions; reviews frontend details.
-
Affinity Diagram
by Owl-Listener · Owl-Listener/designer-skills
Group research observations into themes while preserving their links to the original evidence.
Use an agent that can read the supplied transcripts or notes. Provide participant IDs and the design question; the skill synthesizes existing evidence rather than collecting interviews.
-
Animation Principles
by Owl-Listener · Owl-Listener/designer-skills
Give one transition a clear purpose and a motion profile that supports the interface state change.
Provide the interface's before and after states, trigger and target devices. Use a runnable preview to check timing, interruption and reduced-motion behavior.
-
Card Sort Analysis
by Owl-Listener · Owl-Listener/designer-skills
Analyze participants' card groupings to propose navigation categories and identify ambiguous items.
Provide completed card-sort results, including each participant's placements, labels and the study method. The skill analyzes supplied evidence; recruitment and data collection are separate.
-
Click Test Plan
by Owl-Listener · Owl-Listener/designer-skills
Plan a first-click test to check whether people can identify where to start a task.
Provide the screen or prototype, user goal and recruitment criteria. The skill creates a test plan; running sessions and collecting clicks require a separate study setup.
-
Color System
by Owl-Listener · Owl-Listener/designer-skills
Turn a brand color into a usable set of interface roles and foreground-background combinations.
Provide brand colors, interface roles and accessibility targets. Use actual component states and a contrast-checking tool to evaluate the proposed color combinations.
-
Component Spec
by Owl-Listener · Owl-Listener/designer-skills
Make one component implementable by explaining its behavior as carefully as its appearance.
Provide the component's purpose, visual variants and target platform. The output is a specification for implementation and testing, not an implemented component.
-
Conversational UX
by Owl-Listener · Owl-Listener/designer-skills
Design what the system says, what it expects back, and how it recovers when those two do not match.
Provide supported tasks, channel constraints and confirmation requirements. Voice flows also need device and speech-recognition testing; a written dialog alone cannot establish recognition quality.
-
Dark Mode Design
by Owl-Listener · Owl-Listener/designer-skills
Translate the interface hierarchy into dark surfaces instead of inverting its colors.
Start with an existing light-mode palette, component examples and media assets. Use a rendered interface to check dark-mode contrast, theme switching and image treatment.
-
Data Visualization
by Owl-Listener · Owl-Listener/designer-skills
Choose the graphic that makes the comparison understandable, not the chart that looks most elaborate.
Provide the actual data, field definitions, reader's question and display constraints. The skill guides chart design; data validation and chart implementation remain separate tasks.
-
Design Token
by Owl-Listener · Owl-Listener/designer-skills
Give visual decisions names and relationships that components can reuse.
Provide existing visual values, component examples and naming constraints. The skill defines token names and relationships; export tooling and framework integration require separate setup.
-
Diary Study Plan
by Owl-Listener · Owl-Listener/designer-skills
Capture experiences that unfold between sessions rather than trying to reconstruct them in one interview.
Provide the research goal, participant criteria, time span and feasible recording tools. The output is a study plan; recruitment, check-ins and participant data collection still need to be carried out.
-
Empathy Map
by Owl-Listener · Owl-Listener/designer-skills
Share a user's perspective while keeping direct evidence distinct from interpretation.
An agent that can read your research notes, a defined user or segment, and the design decision the map should inform.
-
Error Handling UX
by Owl-Listener · Owl-Listener/designer-skills
Make failures explain what happened and give people a way forward without losing their work.
An agent that can read the affected flows, known failure cases and the recovery actions your product actually supports.
-
Form Design
by Owl-Listener · Owl-Listener/designer-skills
Specify the path from the first field to a successful submission, including corrections and return visits.
An agent that can read the form brief, the required fields and validation rules, and any existing form screens.
-
Gesture Patterns
by Owl-Listener · Owl-Listener/designer-skills
Design what happens before, during, and after a gesture—not just the final swipe action.
An agent that can read the interaction brief, the target platforms, and the actions available through buttons or menus.
-
Heuristic Evaluation
by Owl-Listener · Owl-Listener/designer-skills
Review interface usability and document the evidence behind each finding, without presenting expert judgment as participant research.
An agent that can inspect the interface, a defined set of user tasks, and relevant domain constraints.
-
Icon System
by Owl-Listener · Owl-Listener/designer-skills
Give icons a consistent construction and naming system instead of collecting unrelated symbols.
An agent that can inspect representative icons, their intended sizes, and the required design and code formats.
-
Illustration Style
by Owl-Listener · Owl-Listener/designer-skills
Give commissioned or recurring illustrations a shared visual language and a clear job in the interface.
An agent that can inspect visual references, product colors, and the screens and sizes where illustrations will appear.
-
Information Architecture
by Owl-Listener · Owl-Listener/designer-skills
Decide where information belongs before deciding what the navigation looks like.
Needs product content and user-task evidence; the source does not require a particular design application.
-
Interview Script
by Owl-Listener · Owl-Listener/designer-skills
Turn research objectives into questions that invite evidence instead of suggesting an answer.
Research goals, participant criteria, and session constraints are the primary inputs; no recording or recruitment service is supplied.
-
Jobs-to-Be-Done
by Owl-Listener · Owl-Listener/designer-skills
Reframe a product decision around what someone is trying to accomplish.
Requires user research or clearly labeled hypotheses; it does not collect interviews itself.
-
Journey Map
by Owl-Listener · Owl-Listener/designer-skills
Connect individual touchpoints into the experience a person actually goes through.
Uses existing personas, research, and analytics; no automatic sentiment measurement is established.
-
Layout Grid
by Owl-Listener · Owl-Listener/designer-skills
Define shared columns, margins and responsive behavior for related page layouts.
No named runtime dependency; requires real content constraints to select useful grid values.
-
Loading States
by Owl-Listener · Owl-Listener/designer-skills
Show what is happening during a wait without misleading people about progress.
Accurate operation timing and progress capabilities must come from the product or implementation team.
-
Localization Design
by Owl-Listener · Owl-Listener/designer-skills
Prepare the interface for language differences, not just translated strings.
Needs target-locale text, fonts, and appropriate language or cultural review; no automatic localization platform is bundled.
-
Micro-Interaction Spec
by Owl-Listener · Owl-Listener/designer-skills
Give engineering the behavior behind a small interaction, not just an animation reference.
No particular motion library is required; target-device testing is still necessary.
-
Motion System
by Owl-Listener · Owl-Listener/designer-skills
Make animation decisions reusable across components and teams.
Requires a token implementation or platform equivalent; the source does not mandate an animation framework.
-
Navigation Patterns
by Owl-Listener · Owl-Listener/designer-skills
Turn an information hierarchy into recognizable navigation controls and clear paths between destinations.
Needs an information hierarchy and task priorities; no navigation component package is provided.
-
Onboarding Design
by Owl-Listener · Owl-Listener/designer-skills
Help new users reach a useful result before asking them to learn the whole product.
Needs product-specific activation definitions and instrumentation for any later measurement.
-
Search UX
by Owl-Listener · Owl-Listener/designer-skills
Design the full retrieval experience, including the searches that fail.
Search services and query instrumentation are external to this guidance.
-
Service Blueprint
by Owl-Listener · Owl-Listener/designer-skills
Reveal the operational work behind what the customer sees.
Needs stakeholder and process evidence; does not access operational systems on its own.
-
Spacing System
by Owl-Listener · Owl-Listener/designer-skills
Choose spacing by relationship and purpose, not by isolated pixel adjustments.
No specific token tool is required; application depends on the existing component system.
-
State Machine
by Owl-Listener · Owl-Listener/designer-skills
Describe what can happen next from every state of a component.
No named state-machine package is required; actual implementation and tests remain downstream work.
-
Theming System
by Owl-Listener · Owl-Listener/designer-skills
Let one component library change appearance without duplicating its structure.
Implementation uses the project's token system or platform equivalents; runtime switching must be built and tested.
-
Typography Scale
by Owl-Listener · Owl-Listener/designer-skills
Create a readable hierarchy from a small set of documented text styles.
Requires the actual font files or available platform fonts for meaningful visual validation.
-
Usability Test Plan
by Owl-Listener · Owl-Listener/designer-skills
Decide what the study needs to learn before scheduling the sessions.
Requires a usable test artifact and access to participants; no recruitment or session platform is supplied.
-
User Flow Diagram
by Owl-Listener · Owl-Listener/designer-skills
Show the routes a user can take to complete one goal.
No diagramming application is required by the source; the chosen output format must preserve clear notation.
-
User Persona
by Owl-Listener · Owl-Listener/designer-skills
Give product decisions a consistent user reference without inventing a customer story.
Requires genuine user evidence for research-grounded claims; supplied product URLs may need separate investigation.
-
Visual Hierarchy
by Owl-Listener · Owl-Listener/designer-skills
Choose what should be noticed first, then make the composition support that choice.
Needs real content and rendered review; no visual scoring tool is bundled.
-
Wireframe Spec
by Owl-Listener · Owl-Listener/designer-skills
Make a low-fidelity screen precise enough to discuss behavior and content.
No particular wireframing tool is required; a usable brief must include content and state information.
-
theSVG
by glincker · glincker/thesvg
Retrieve an existing brand mark or service icon in the form your project needs.
Needs network access to the registry or CDN. React, CLI, Figma, and MCP integrations are optional surfaces rather than prerequisites for simple URL retrieval.
-
Image to Editable PPT
by ningzimu · ningzimu/image-to-editable-ppt-skill
Convert visual slides into a structured deck instead of placing screenshots behind editable text.
Requires the editppt CLI, suitable image-generation capabilities, and multi-agent execution for multi-page runs. PaddleOCR is an optional configured aid; offline hints have a different quality profile.
-
accessibility
by Addy Osmani · addyosmani/web-quality-skills
Audits and improves web accessibility following WCAG 2.2.
-
Landing Page Design
by elayadesign · elayadesign/ai-design-skills
Connect the page's argument and its visual rules around a single intended conversion.
No single application runtime is required; implementation examples reference Tailwind values, IntersectionObserver, and optional motion tooling.
-
Anything2Explainer
by Vincentwei1021 · Vincentwei1021/anything2explainer
Build an original, source-backed explainer with narration and a deliberately consistent visual language.
Requires the supplied Remotion/Node project, Python helper scripts, rendering capability, and a suitable TTS route. The documented Chinese and English defaults use different voice engines; multi-agent construction is part of the described workflow.
-
Logo Design
by kaankiziltug · kaankiziltug/logo-design-skill
Build and test the mark before expanding it into a full brand kit.
The source describes dependency-free Python helper scripts, while actual image rendering needs an available renderer such as CairoSVG, Inkscape, Chrome, or another supported option.
-
Holo Card Studio
by EverettFish · EverettFish/holo-card-studio
Choose a foil treatment or a full-image angle-dependent transformation before building the card.
Requires image generation, Python, Blender, and a web runtime for the delivered viewer. The source mentions local dependency installation during the pipeline.
-
Dream Loop
by achimala · achimala/dream-loop
Establish the desired screenshot first, then use it as the target for implementation.
Needs a target image, or image-generation access if none is supplied. Detailed tool requirements depend on the selected referenced workflow and were not established from the entry alone.
-
motion-design
by LottieFiles · LottieFiles/motion-design-skill
Applies motion design principles: timing, easing, choreography, and animation fundamentals.
-
user-research-cookiy
by Cookiy · cookiy-ai/user-research-skill
End-to-end user research assistant: qualitative and quantitative, guides and plans.
-
material-3
by hamen · hamen/material-3-skill
Implements Material Design 3 tokens, components, and adaptive layouts.
-
huashu-art-motion
by alchaincyf · alchaincyf/huashu-art-motion
Build a short film from an art direction, not a sequence of slide transitions. Huashu Art Motion combines code-drawn scenes, character sprites and motion review in a local rendering workflow.
Local rendering requires uv, FFmpeg and Playwright Chromium. Supply existing character frames or use a separate image-generation workflow for human-like sprites; audio and fonts need their own assets and permissions.
-
Academic PPTX
by Gabberflast · Gabberflast/academic-pptx-skill
Plan an academic presentation around its argument, evidence and slide titles; a separate tool creates the PPTX file.
Requires a separate PPTX Skill. The entry lists markitdown, PptxGenJS, LibreOffice, and Poppler as implementation-related dependencies.
-
design-motion-principles
by Kyle Zantos · kylezantos/design-motion-principles
Purposeful interaction motion, or a review of existing animation, after Kowalski, Krehel, and Tompkins.
-
Hue
by dominikmartn · dominikmartn/hue
Create a reusable design skill from visual references, design rules and tokens.
Needs file access, suitable reference-inspection tools, Node for validate.mjs, and browser review for visual confidence. Tool availability affects extraction fidelity.
-
Power Design
by ItsssssJack · ItsssssJack/power-design
Apply a brand system to an HTML page or slide deck, using layout rules appropriate to the chosen format.
URL-based brand extraction needs Firecrawl. Default outputs avoid external JavaScript dependencies, while fonts or image assets may still require their own availability and rights.
-
color-expert
by David Aerne · meodai/skill.color-expert
Color naming, theory, spaces, palettes, ramps, gradients, conversions, and contrast.
-
AntV G2 Chart
by antvis · antvis/chart-visualization-skills
Create charts with AntV G2 v5 using its marks, encodings, scales and interactions.
Requires @antv/g2 v5 or the documented browser bundle. Optional reference retrieval uses the external AntV context service.
-
AntV G6 Graph
by antvis · antvis/chart-visualization-skills
Visualize network data with AntV G6 v5, including graph layouts and interactive exploration.
An application using @antv/g6 v5, a browser graph container, and network access if the documented reference-retrieval service is needed. Verify APIs against the installed release.
-
AntV X6 Editor
by antvis · antvis/chart-visualization-skills
Build a node-and-edge diagram editor with AntV X6, including connections, editing behavior and required plugins.
Use the documented @antv/x6 v3 API with a browser canvas and the required plugins. Fetching additional references needs network access. Check examples against your installed version.
-
chart-visualization
by AntV · antvis/chart-visualization-skills
Renders data as bar, line, pie, radar, Sankey, mind-map, and flow charts through the AntV API.
Needs: AntV API
-
infographic-creator
by AntV · antvis/chart-visualization-skills
Creates infographics from a given text.
Needs: AntV API
-
ux-writing
by content-designer · content-designer/ux-writing-skill
Writes and reviews interface copy with tone guidance, editing passes and usability checks.
No matching skills.
02 How to choose and start
How to Choose a Design Skill and Run It in Claude Code or Codex.
Pick by task and output, check what the skill needs, then install it in your agent and run one real task. The same three steps work in Claude Code, Codex, Cursor and OpenCode.
- 1
Choose the Task, Not the Skill Name
Filter by what you need to make — a landing page, a deck, a brand system, a review. Compare the one-line use and the output type before you compare stars.
CheckTwo or three candidates whose output matches the file you actually need.
- 2
Check What the Skill Needs
Open the source and read the requirements. Some skills only work with a specific tool such as Google Stitch, HyperFrames, or GSAP; some need an image-generation service. The catalog lists these under Needs when the SKILL.md states them.
CheckThe tool or service the skill names is available in your project before you install.
- 3
Install It in Your Agent and Run One Real Task
Install the skill the way its author documents it — for most repositories that is one
npx skills addcommand — then run one real task with your own content in Claude Code, Codex or your agent. Copying a command is not a successful install until the agent recognizes and uses the skill.CheckThe agent lists the skill by name and its output changes when you invoke it.
03 Run them in OpenDesign
Add Skills to Your Design Workspace.
OpenDesign works with coding agents and SKILL.md instructions. A skill may also depend on external software, provider credentials or a particular agent runtime; importing its instructions does not supply those dependencies.
- 38
Recorded as Bundled
The catalog’s In OpenDesign tag records an earlier product check (2026-08-28), including frontend-design and the GSAP skills. Check availability in your installed version; this source review does not verify the current bundle.
- ↗
Import a Skill From Its Link
Use each card’s Source link to inspect the entry file, then follow the repository and your installed version’s setup instructions. Keep required scripts, references and assets with the skill; importing a single instruction file may not be enough.
- ›_
Runs with the Agent You Already Use
Choose an agent and runtime that support the skill’s required tools. Reading its instructions is not the same as executing its workflow: local software, connectors, credentials and any separate services still need to be available.
04More design resources
Keep Exploring
A skill gives your agent instructions for a task. Use these resources to choose a visual direction, plan a website, or find a reference worth exploring before you start building.
-
Design System
Explore color, typography, spacing, and component references to define a consistent visual direction. Pair that direction with a skill when you want your agent to follow a clear design brief.
-
AI Website Builder
See how to approach a website project with AI, from an initial brief to page structure and visual refinement. Use it alongside a skill when the goal is a complete website.
-
Showcase
Browse community projects for ideas on layout, visual style, and presentation. Choose a reference, identify what makes it work, and use those observations to make your next skill prompt more specific.
OpenDesign
Run Design Skills with the Coding Agents You Already Use.
OpenDesign is an open-source, local-first design workspace. It turns coding agents into a design engine for prototypes, web pages, slides, and HTML video — real runnable files, on your own machine, with your own keys.
These skills are independent projects with different licenses and runtime requirements. Bundled entries are marked in the catalog. Review the upstream setup and terms before importing other skills.
